Timothy Armes' blog

A Photographer in this digital age will inevitably encounter the confusing world of meta-data, and with it terms such as IPTC, EXIF and XMP.  He or she may begin to have a feeling for what these all mean, but often they’ll remain somewhat of a mystery, a computery-technical-thingy that isn’t very relevant to their photography.

In my last blog post I talked about my new plug-in, PLUS for Lightroom, that’s used to embed licensing rights information into the XMP meta-data of an image.  I received a couple of questions about XMP so I thought I’d explain the basics of meta-data for the benefit of those who are just starting to get to grips with the digital workflow.

Meta-Data

Meta-data is just data about data.  Information about information.

Huh?

A concrete example may help.  If the data is a photo, then the meta-data will include information about that photo, such as the time it was taken, the photographer, the shutter speed and aperture, etc.  This meta-data isn’t part of the captured image itself, but it’s information about that image. Conceptually, it’s as simple as that.

The usefulness of meta-data becomes apparent when the photographer starts to appreciate the importance of a good workflow, and it becomes essential knowledge when photographs are passed between different people in a professional situation.  As a very basic example, a photographer will wish to ensure that a copyright line is carried with each of his or her images.

Meta-data can be stored in a variety of places. If you write down some notes about your photo and stick them on a pile on your desk, that’s meta-data. Obviously that would be far from ideal though.  A better solution is to use a computer to help us with this task, and there are tools designed for photographers, such as Lightroom, Photo Mechanic or Aperture that are designed to do just that. These programs usually maintain a database in which meta-data about each photo (such as a description, it’s location on the hard drive and a 5 star rating) is stored.  Since the metadata is centralised into one database it’s very quick to run searches and filters, and to do other cool things that will help your workflow.

There’s another option for meta-data storage too.  The meta-data can be embedded inside the file with which it’s associated, which is to say that it’s stored inside the file (the JPEG, TIFF, etc) in such a way that it can be retrieved by a capable application.  This method has the advantage that the meta-data is actually attached to the file, so if the file moves or gets sent somewhere by email, then the meta-data will travel with it.  The disadvantage of this approach is that it’s less efficient if you need to do a search; if the computer needs to open every file to read its meta-data then this would slow down meta-data searches significantly. For photos stored elsewhere (such as those archived on a CD), then you would be required to insert the CD just to complete your search request.

In reality we use a combination of approaches.  Some meta-data is stored in the file, and some within the database software.  Most often however, the same data is stored in both places, so that searches are efficient whilst important information is guaranteed to be kept securely with the file itself.

EXIF & IPTC

The terms EXIF and IPTC refers to two different and complimentary meta-data standards.

EXIF, stands for Exchangeable Image File data.  It’s a standard for storing information concerning the camera’s settings at the moment the picture was taken.  The EXIF meta-data contains things such as the shutter speed, aperture, flash settings, exposure compensation, focus point, colour space, focal length, time and date, etc.  The camera stores this meta-data with every picture that you take.

The IPTC is the International Press Telecommunications Council. In 1970 the IPTC defined a standard for meta-data that was designed for improving the digital workflow within the press world. IPTC meta-data contains information that is added to a photo by photographers and editors, such as the copyright information, image descriptions, descriptive keywords, etc.  Over time the standard has evolved, and there are a host of possible fields that may be filled in depending on the needs of the organisations involved.

Both of these standards have had to evolve over time, and this hasn’t been a straight-forward task.

The IPTC standard started life as a text-based standard for news related articles and gradually expanded to meet the needs of television and radio industries.  It wasn’t until 1994 that Adobe would include the ability to store IPTC data (or “IPTC headers”) into its digital image file formats. These traditional IPTC headers  are no longer appropriate for modern usage. Lacking proper support for non-English characters (even accented characters are not well supported) foreign languages may be impossible to incorporate.  Furthermore, the standards defining the location of the headers inside various file formats aren’t as strict as they should be.  You may be able to read the headers in one application but not another. The CrossIPTC tool is an example of a tool that attempts to convert IPTC data for use between platforms.

EXIF meta-data has it’s own problems.  While there are a lot of standardised fields, camera makers have also added their own proprietary information.  Since this information and the way in which it’s stored changes from one make to another, imaging applications may not be able to read and display all of it.  Worse still, saving a modified version of a file may result in a loss or corruption of this data, since the application isn’t even aware that it’s there.

XMP

With all this past experience came the opportunity to create a new all-encompassing meta-data standard.  Creating a new standard is a hard thing to do – people don’t like change.  To create a new standard requires a huge amount of time, money and drive.  It also requires the support of the industry. In 2001  Adobe stepped up to the challenge and created XMP.

XMP, the Extensible Meta-data Platform, provides a standard for storing meta-data, any meta-data, with the associated document. Adobe used their weight within the industry to gradually push XMP to the forefront.  All of the Adobe Creative Suite products provide support for XMP meta-data, giving virtually every digital imaging professional the ability to work with XMP.

Structured upon open standards that have been proved (XML), the data is stored in such a way that any of the world’s languages may be used in any text.  The standard is well defined and completely open, so it’s possible for all developers to write conforming applications.

XMP is, by definition, extensible.  If an organisation needs to store a piece of meta-data in its files that isn’t defined in the standard, then this information can be added to the XMP meta-data such that can be read and written by XMP aware imaging applications¹.  The applications don’t need to be changed to cope with the addition, and there shouldn’t be any way in which two proprietary meta-data blocks can be confused with each other.

Finally, one of XMP’s greatest strengths is that it defines how this information is stored in a particular file type. For file types that don’t support meta-data inclusion, XMP meta-data can still be stored in an associated file (known as a sidecar file) which should be moved and copied with the original.

Structurally, XMP is based around schemas.  A schema is simply a collection of related data, and there are standardised XMP schemas for the most common types of meta-data.  This brings the advantages of XMP to the well known meta-data standards that are used throughout the world today.  For example, there are schemas for both EXIF and IPTC, providing a robust system for including these standards, and giving the them freedom to develop.

It’s thanks to the extensibility of XMP that the PLUS coalition have been able to develop the PLUS standard for embedding usage rights information into image files.

Multiple Meta-Data

You may be wondering what happens if you include XMP meta-data in a file that already contains traditionally encoded IPTC and EXIF data.

It’s a great question and the answer is that it depends on your application.  If you’re running an older application that isn’t XMP aware, then it’ll just read the old IPTC and EXIF data as usual.  A newer application may see the XMP, and use only that, assuming that it overrides any older forms of meta-data, or it may offer you access to everything.

When images are saved newer applications may convert the IPTC and EXIF data into the new XMP schemas. The application may also try to simultaneously write the meta-data in the old format so that older applications can read it, however there are limitations to this approach.  Over time the old formats will be dropped and only XMP will be supported, removing all ambiguity.

XMP is a great creation.  Embrace it!

1. Within Photoshop, it’s even possible to design custom panels that will be displayed in the File Info box along with the other XML meta-data. Adobe supply document that explain how this is done. [↩]

This entry was posted on Monday, November 9th, 2009 at 11:12 am and is filed under Business Practises. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
The short URL for this post is http://tarm.es/b/eaz.